Has anyone in the goTenna community looked at distributed webpages? Since the bandwidth is so hard, maybe try an IPFS protocol:
1 Like
Another example of this distributed information structure is a process known as “seeding”. This is where every web page you visit is stored on your device so you can serve that to the next guy. The web browser “Beaker” Does exactly that:
2 Likes
Maybe in a few years. You mentioned the difficulty of providing enough bandwidth, but you’re talking about the internet and what the narrator described as storage outpacing the capacity growth of bandwidth failing to keep up with demands made to fill that storage.
Mesh networking has its own bandwidth limitations, as well as considerable strengths vs the internet. I could see it assisting with a solution like this particularly in dealing with the last mile problem between user and fiber. But this would assume that traffic was compressed, tailored and filtered to fit the form-factor provided by the very limited bandwidth available on mesh while still keeping it in a physical size and mass given current and near-term technology.
Particularly troublesome here would be the hidden node problem, since one would assume with distributed data solutions that IPFS seems to provide there would be large volumes of requests for pieces of documents would likely produce a significant overhead burden on mesh networks.
Well worth a read and as an aid to better understanding how mesh works, see Ram Ramanathan’s article:
Mesh networking explainers by goTenna’s Chief Scientist!
1 Like
No the idea is that you can retrieve an image or even an entire webpage from your neighbor instead of taking up the bandwidth of all your neighbors between you and the host. It saves bandwidth…
1 Like
IIRC, experiments with slow scan TV to send a very low rez images over mesh resulted in a time measured in hours to complete the download. Between the low resolution and the download time, I really doubt that internet users would find that acceptable.
I think you’re failing to appreciate the severe bandwidth limitations that are inherent to mesh networks. Sure, if goTenna was as ubiquitous as cell phones, then maybe that image could be broken down into smaller pieces and a whole bunch of close by goTenna Meshes would be able to speed that up some in that way. Even then, I suspect the results would be less than optimal.
Now if you want to base such a network on cell phones, that would certainly help lessen bandwidth demands on the internet backbone, but that would require a continuing reliance on highly centralized infrastructure. goTenna is an effort that has its basis a preference for decentralized solutions. It doesn’t sound like a good fit to me, but maybe you can explain how to overcome this significant hurdle.
Perhaps if we dropped down to a LYNX style text-only, bare minimum webpage in conjunction with your locally seeded webpages? Unfortunately, Mike is right, the image data alone would cripple the network.
I came in with the assumption that the network already works…does it not already work? Because if it already works then this will help make it better…if it doesn’t already work then obviously we gotta work on that first.
It works fine for text messaging.
What you’re suggesting is putting a ton of bandwidth into a 5 pound bag.
That’s a little too extreme for the Gotenna Mesh protocol. Think more along the lines of Twitter without pictures. Also you need to work within the devices capability. The Gotenna itself doesn’t have enough memory to handle a bunch of webpages and how many do you think the average user is willing to store in their phone?
Now think about the batteries in the phone and the Gotenna. Mobile isn’t quite ready for this idea.